An Energy-Efficient Load Balancing Algorithm for Virtual Machine Environments to Perform Communication Type Application Processes

In fault-tolerant information systems, application processes have tobe reliably performed in presence of server faults. Server clustersystems are widely used to realize not only high performance but alsofault-tolerant service. In order to reliably perform applicationprocesses, multiple replicas of each application process are performedon multiple servers in a server cluster system. On the other hand, thelarger amount of electric energy is consumed in a server clustersystem since multiple replicas of each application process areperformed on multiple servers. The redundant delay time based (RDTB) algorithm is discussed to select multiple servers to redundantlyperform each application process in a homogeneous server cluster sothat the total energy consumption of the server cluster can bereduced. In a homogeneous cluster, every server follows the samecomputation model and power consumption model. On the other hand, servers follow different types of computation models and powerconsumption models in a heterogeneous server cluster. Hence, a loadbalancer cannot precisely estimate the processing load and energyconsumption of each server in a heterogeneous server cluster by theRDTB algorithm. In this paper, we propose an improved RDTB (IRDTB) algorithm to select multiple servers for redundantly performing eachapplication process so that the total energy consumption of a servercluster and response time of each process in a heterogeneous servercluster can be reduced. We evaluate the IRDTB algorithm compared withthe RDTB algorithm in a heterogeneous server cluster. In theevaluation, we show the average total energy consumption of aheterogeneous server cluster in the IRDTB algorithm can be reducedmaximum 44% of the RDTB algorithm.

Server cluster systems equipped with virtual machines are widely used to realise scalable and high-performance computing systems. In order to satisfy application requirements like response time, processing loads of virtual machines have to balance with one another in a server cluster. In addition to achieve the performance objectives, the total electric energy of a server cluster to perform application processes has to be reduced. In this paper, the Energy Consumption Laxity Based (ECLB) algorithm is proposed to allocate computation type application processes to virtual machines in a server cluster so that the total electric energy of the server cluster and response time of each process can be reduced. We evaluate the ECLB algorithm compared with basic Round-Robin (RR) algorithm. Evaluation results show the average total electric energy of a server cluster and average response time of each process in the ECLB algorithm can be more reduced than in the RR algorithm.

The total electric energy consumption of a server cluster to perform application processes on multiple virtual machines has to be reduced to realize energy-efficient information systems. In our previous studies, the ECLB (energy consumption laxity based) algorithm to allocate computation type application processes (computation processes) to virtual machines in a server cluster is proposed so that the total electric energy consumption of the server cluster can be reduced. Here, each virtual machine is bounded to one thread on a CPU in a server and computation processes are performed by using only one thread bounded to the virtual machine even if some threads are not used in the server. In this paper, the improved ECLB (IECLB) algorithm is proposed to furthermore reduce the total electric energy consumption of a server cluster by allocating unused threads to each virtual machine to perform computation processes. We evaluate the IECLB algorithm in terms of the total electric energy consumption of a server cluster compared with the ECLB algorithm.

Various types of scalable and high performance computing systems are developed to realize distributed applications. In order to realize not only scalable but also high performance information systems, serverclusterc systems are widely used. In server cluster systems, the larger electric energy is consumed since application processes are performed on the larger number of servers. The improved delay time-based (IDTB)algorithm is discussed to select a server for each request process in a homogeneous server cluster so that the total energy consumption of the homogeneous server cluster to perform application processes can be reduced. In a homogeneous cluster, every server follows the same computation model and power consumption model. In heterogeneous cluster, servers follow different types of computation models and power consumption models. In this paper, we propose an extended IDTB (EIDTB) algorithm to reduce the total energy consumption and response time of each process in heterogeneous server cluster. We evaluate the EIDTB algorithm compared with the IDTB algorithm in a heterogeneous server cluster. In the evaluation, we show the average total energy consumption of a heterogeneous server cluster in the EIDTB algorithm can be reduced maximum 57% of the IDTB algorithm.

The electric power consumed by servers has to be reduced in a cluster in order to realize eco society. We take a macro level approach to reducing the total electric energy consumption of servers to perform application processes in a server cluster. Servers are now equipped with multi-core CPUs. In this paper, we discuss a multi-level power consumption (MLPC) model of a server with a multi-core CPU. Here, the power consumption of a server depends on the number of active cores and active threads where at least one application process is performed. We also discuss a multi-level computation (MLC) model which gives the expected execution time of a process which is concurrently performed with other processes on a server with a multi-core CPU. Based on the MLPC model and the MLC model, we discuss an energy-aware (EA) algorithm to select a server for each process requested by a client in a cluster so as to reduce the total electric energy consumption while satisfying deadline requirements of the processes. We evaluate the EA algorithm and show not only the total energy consumption but also the average execution time of each process is reduced in the EA algorithm compared with the round-robin (RR) and random (RD) algorithms.

It is critical to reduce the electric energy consumption of servers in a cluster. In this paper, we discuss a migration approach to reducing the electric energy consumption of servers where a virtual machine with application processes migrates to a more energy-efficient server. In this paper, we consider a homogeneous cluster. We newly propose an ISEAM2H algorithm. Here, a virtual machine on a host server is selected to perform a process issued by an application and a guest server to which a virtual machine migrates is found so that the total electric energy consumption of servers can be minimized. In the evaluation, we show the total electric energy consumption and active time of the servers and the average execution time of processes can be reduced in the ISEAM2H algorithm.

: The North American Aerospace Defense Command (NORAD) and United States Northern Command (USNORTHCOM) (N-NC) crew training has been hindered by an inability to conduct dynamic training and exercises on multiple Command and Control (C2) systems. There was no common simulation injector because of stove-piped acquisitions and legacy interfaces that were incompatible with the Live-Virtual-Constructive Toolkits. The N-NC Joint Training and Exercise Directorate could not afford traditional replication or emulation of all C2 systems and their data sources, nor the inevitable sustainment costs. This paper presents a cost-effective solution to provide dynamic scenario injection into multiple C2 systems: leveraging server and desktop virtualization technology described in previous I/ITSEC papers. The virtualization process transforms stand-alone systems into functionally equivalent virtual machines (VMs). Server virtualization technology lets multiple VMs run as guests on a single host, and a host can support VMs running different operating systems. This allows entire processing strings, distributed throughout North America, to be converted into VMs on a single server. Because the VMs inherit the fidelity of the actual processors, their outputs are as authentic as the operational systems. These VMs feed processed simulation event data into actual C2 systems or equivalent VMs. Future operational system upgrades can be virtualized and then replace existing VMs without changing this infrastructure. Desktop virtualization technology allows users to run multiple VMs in separate windows on a common display. NNC exploited desktop virtualization to simplify the trainee and model operator's workspaces. They can view and manage multiple VMs with one monitor, keyboard and mouse (controlling simulations, lower echelon processing and operator interaction, and viewing C2 workstation displays).

In our previous studies, the dynamic virtual machine migration (DVMM) algorithm is proposed to reduce the total electric energy consumption of servers by migrating virtual machines to more energy-efficient servers. Here, virtual machines are dynamically created and dropped depending on number of processes. However, the more number of virtual machines are created, the more number of virtual machines are dropped. Servers consume the electric energy to drop virtual machines. In this paper, we propose an IDVMM (improved DVMM) algorithm to decrease the number of virtual machines to be dropped. In the evaluation, we show not only the total electric energy consumption and active time of servers but also the average execution time of processes can be reduced in the IDVMM algorithm compared with other algorithms.

As server consolidations based on the virtualization techniques become popular and cloud services continue to grow rapidly, more and more data centers are being built to accommodate a number of virtual clusters running various workloads. Since these virtual clusters often share the resources provided by physical machines (PMs), it is more likely that the interferences between virtual machines (VMs) affect the performance of applications running on top of the virtual clusters. While a lot of studies have proposed different virtual machine placement algorithms to investigate this issue, the problem caused by network performance variability still remains as a challenging issue. Since they usually ignore the CPU overhead to process the communications between VMs, the network bandwidth allocated to a VM cannot be fully utilized when a PM has not enough CPU resources to cover the CPU overhead for VM networking functions. This results in unpredictable application performance running on the virtual clusters. This paper proposes a virtual machine placement algorithm that considers the CPU overhead incurred to reserve network bandwidth in a virtualized data center environment. In order to decide the CPU overhead necessary to guarantee the network bandwidth allocated to a VM, a performance model based on standard linear regression using the data collected from a real environment is used. By comparing the amount of CPU resource available in the driver domain with the CPU overhead obtained from the performance model, the proposed algorithm decides whether the network bandwidth for the VM can be provided or not and selects an appropriate location for VM placement. The benchmarking results show that the proposed algorithm guarantees the network bandwidth allocated to each VM without violations when the CPU resources are shared by multiple VMs.

This paper addresses the problem of simultaneously migrating a group of co-located and virtual machines (VMs), i.e, VMs executing on the same physical machine. We refer to such a mass simultaneous migration of active VMs as live gang migration. Cluster administrators may often need to perform gang migration for load balancing, system maintenance, or power savings. Application performance requirements may dictate that the total migration time, network traffic overhead, and service downtime, be kept minimal when migrating multiple VMs. State-of-the-art migration techniques optimize the migration of a single VM. In this paper, we optimize the simultaneous migration of multiple co-located VMs. We present the design, implementation, and evaluation of a de-duplication based approach to perform concurrent migration of co-located VMs. Our approach transmits memory content that is identical across VMs only once during migration to significantly reduce both the total migration time and network traffic. Using the QEMU/KVM platform, we detail a proof-of-concept prototype implementation of two types of de-duplication strategies (at page level and sub-page level) and a differential compression approach to exploit content similarity across VMs. Evaluations over Gigabit Ethernet with various types of VM workloads demonstrate that our prototype for gang migration can achieve significant reductions in both network traffic and total migration time.

In elastic cloud computing environment, multiple virtual machines may reside in the same physical machine for services consolidation. For the same residential guest domains or multi-tiered hosting services, the inter-domain communications are complex and frequent. However, traditional inter-domain communications are conducted through the virtual network interfaces of both sending and receiving virtual machines. Such TCP/IP networking communications involve lengthy communication path and frequent kernel interactions and thus downgrade the I/O performance of virtual machines. In this paper, we present an optimized inter-domain communication approach based on shared memory to improve the communication performance of multiple virtual machines residing in the same physical machine. In our approach, the sending data is shared in memory, not transferred via TCP/IP stack. The communication data are directly mapped into user space of the virtual machine and therefore it reduces useless system calls and context switches. The memory is shared via a customized shared device kernel module with bidirectional event channel resided in both communicating virtual machines. We use state flags in circular buffer to reduce wait and notify operations and thus reduce the system calls. Experiments show that our proposed approach can provide 5 time higher throughput and 2.5 times less latency compared to TCP/IP communication path.

In order to realize green society, we take the process migration approach to reducing the total electric energy consumption of clusters by migrating virtual machines. In our previous studies, the DM (Dynamic Migration) algorithm is proposed where virtual machines are dynamically resumed and suspended and a group of virtual machines on a host server migrate to a guest server so that the total energy consumption of the host and guest servers can be reduced. In this paper, we newly propose a DGM (Dynamic Group Migration) algorithm. Here, there is a seed virtual machine on each server. A virtual machine is created as a clone of the seed virtual machine so that the number of processes on each virtual machine is kept fewer. In the evaluation, we show the total electric energy consumption of servers can be more reduced in the DGM algorithm compared with other algorithms.

Processes on virtual machines can migrate from a host server to a guest server by migrating the virtual machines. In this paper, we propose a DMMV (Dynamic Migration of Multiple Virtual machines) algorithm where virtual machines migrate from a host server to a more energy-efficient guest server. Here, virtual machines are dynamically suspended and resumed depending on the number of processes. In addition, one or more than one virtual machine is selected to migrate from a host server to a guest server. The number of virtual machines to migrate is decided so as to minimize the electric energy consumption of the host and guest servers. In the evaluation, we show not only the total electric energy consumption and active time of servers but also the average execution time of processes can be reduced in the DMMV algorithm compared with other algorithms.

Optimizing dynamic survivability and security of replicated data in cloud systems under co-residence attacks

Infrastructure-as-a-Service (IaaS) clouds have become more popular enabling users to run applications under virtual machines. This paper investigates the energy-aware virtual machine (VM) allocation problems in IaaS clouds along characteristics: multiple resources, and fixed interval times and non-preemption of virtual machines. Many previous works proposed to use a minimum number of physical machines, however, this is not necessarily a good solution to minimize total energy consumption in the VM placement with multiple resources, fixed interval times and non-preemption. We observed that minimizing total energy consumption of physical machines is equivalent to minimize the sum of total completion time of all physical machines. Based on the observation, we propose EMinRET algorithm. The EMinRET algorithm swaps an allocating VM with a suitable overlapped VM, which is of the same VM type and is allocated on the same physical machine, to minimize total completion time of all physical machines. The EMinRET uses resource utilization during executing time period of a physical machine as the evaluation metric, and will then choose a host that minimizes the metric to allocate a new VM. In addition, this work studies some heuristics for sorting the list of virtual machines (e.g., sorting by the earliest starting time, or the longest duration time first, etc.) to allocate VM. Using the realistic log-trace in the Feitelson's Parallel Workloads Archive, our simulation results show that the EMinRET algorithm could reduce from 25% to 45% energy consumption compared with power-aware best-fit decreasing (PABFD) [1]) and vector binpacking norm-based greedy algorithms (VBP-Norm-L1/L2 [2]). Moreover, the EMinRET heuristic has also less total energy consumption than our previous heuristics (e.g. MinDFT and EPOBF) in the simulations (using same virtual machines sorting method).